Ram Purge?


1812

Recommended Posts

Weird/stupid question, but is there a way to purge cached data from ram?

 

And before anyone asks "Why would you want to do that?" I'll answer: I have a very use specific case that does not affect 98% of users.

 

One of my servers is sitting on a large amount of ram (north of 82GB) Why do I need so much? Well, multiple vm's that do photo/video editing, plex transcoding, crashplan, etc... It's a fantastic all-in-one-at-the-same-time box. It does draw a little bit of power given all the pcie cards/etc, but much less than breaking all that out into separate machines, or even just 2 machines.

 

The issue I perceive: if I have 32 GB available for plex/network transfers/etc, unRaid keeps that data even after it is used, keeping the ram active. My server, like many others, has the ability to lower the ram power usage  when not active (ie: empty) saving anywhere between 4-40 watts of power (lots of sticks in the server.) I know this through my own testing on fresh boot with vm's loaded and 40% free ram vs a few hours later with ram full. 4 watts isn't much, but if I saved 40 watts every night, that could add up to a nice steak dinner at the end of the year.

 

I've done a little reading and know there are terminal commands to empty cache buffers. But that starts to get over my head a little. I suppose those could be put in user scripts?

 

Restarting plex dump the cache I believe, so I could have that auto-restart on a schedule?

 

Any thoughts? Am I just being too picky?

 

 

 

Edited by 1812
Link to comment
Just now, Squid said:

Maybe I'm thinking this wrong, but if the board does do that wouldn't the 4-40Watts savings be at ~1.3 Volts.  Once that's all converted to the actual power draw at the outlet, you'd be saving a dollar at the end of the year.  (Nice deal on a steak though  :D )

 

 

I wish your math was right!!! Then it would only cost me 7 dollars a year to run the 40 thread 84gb server. At which point I wouldn't even think about looking at this as an option.

 

I measured power differences on the total power of the server via management software. Comparatively, it is the difference between a low wattage incandescent light bulb being on all the time or not. It's really not a major thing I suppose, but something I was thinking about this morning while upgrading it's twin on steroids. 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.